Journal of Microbiology, Biotechnology and Das et al. 2012 : 2 (1) 162-174 Food Sciences

162

REGULAR ARTICLE

GENOTYPING OF CLOSTRIDIUM PERFRINGENS FROM FRESH WATER FISH

AND FISH PICKLES

Arunava Das*, Adarsh Jain

Address: Department of Biotechnology, Bannari Amman Institute of Technology,

Sathyamangalam -638401, Erode District, Tamil Nadu, India,

Phone number: +91-9751882590

*Corresponding author: drarunavadas@rediffmail.com

ABSTRACT

This study aims to evaluate the genotypes of Clostridium perfringens in fish and fish

based products from Tamil Nadu and Kerala states of India. A total of 301 samples consisting

intestinal contents of freshwater fish (234) from various dams, freshwater lakes, ponds, retail

shops and markets and fish pickles (67) obtained from randomly selected retail shops and

supermarkets were investigated. Bacterial isolations, identifications and phenotypic

characterization of virulence factors were carried out as per standard microbiological

procedures. Genotyping of the C. perfringens isolates were done by amplifying four major

lethal toxin genes namely- alpha toxin gene (cpa), beta toxin gene (cpb), epsilon toxin gene

(etx), iota toxin gene (iA) in a Thermal Cycler. Isolates were also screened for the presence of

enterotoxin gene (cpe) and beta2 toxin gene (cpb2) by single step PCR. Biochemical tests and

phenotypic determination of virulence factors tentatively identified 82 (27.24%) isolates of C.

perfringens. In PCR assay, all 82 (100%) isolates harbored cpa toxin genes of C. perfringens,

however, 65 (79.26%) isolates also carried additional cpb2 toxin genes. None of the isolates

were found positive for beta, epsilon, iota and enterotoxin genes. Genotyping of the 82

isolates by PCR revealed that all the isolated bacteria were belonged to C. perfringens type A

and both cpa and cpb2 toxin genes were prevalent among the isolates of C. perfringens type

A, impending the risk of pathogenicity to human via freshwater fish and fish pickles.

JMBFS / Das et al. 2012 : 2 (1) 162-174

163

Keywords: Clostridium perfringens, alpha toxin gene, beta2 toxin gene, fish, fish pickle

INTRODUCTION

Fish production in India has increased multi fold since India’s independence in 1947.

According to the report of UNFAO (2011), fish output in India doubled between 1990 and

2010. As food items, fish and fish based products contribute to country’s economy and

people’s well being by meeting the dietary protein needs of the common man. This is due to

the easy availability of fish in the river based states and plains of India. Due to their high

perishable characteristic, fish and fish based products easily get spoiled and contaminated by

the food borne pathogens particularly bacteria if not properly stored. When such contaminated

foods are consumed either raw or inadequately cooked by human population, there is a

detrimental impact on public health.

According to the annual report of DAHDF (2009), Kerala and Tamil Nadu states of

India are ranked the 4th and the 5th respectively among the leading fish producing states in

India. Presently, there is an increasing demand of locally processed fish and canned fish

products like fish pickles in these two states of Southern India. Due to existing poor storage

facilities in retail shops or even in local market, the risk of food borne infection is ever

present. Among the food borne pathogens that thrives in high protein and canned food items,

Clostridium perfringens always comes foremost.

C. perfringens is ubiquitous, anaerobic, gram positive rod shaped bacteria which

causes numerous enteric diseases caused by food infections to human, wildlife and domestic

animals (Songer, 1996). C. perfringens thrives in high-protein foods of animal origin such as

meat and meat products, meat dishes, stews, soups, gravies, and milk. On the basis of

production of four major toxins (alpha, beta, iota and epsilon), isolates of C. perfringens are

classified into five genotypes A to E (Songer and Meer, 1996). Along with four major lethal

toxins, enterotoxin and beta2 toxin produced by types of C. perfringens also considered as

important toxins for enteric diseases in domestic animals (Smedley III et al., 2004; Sting,

2009). However, it is not so easy to differentiate between different strains of C. perfringens

from clinical cases. Therefore, PCR has been used to detect the presence of toxin genes and to

identify the specific strains of C. perfringens (Yamagishi et al., 1997).

In last five to six years, the detection of C. perfringens type A from black leg in cattle

(Shome et al., 2006a), gangrenous dermatitis in broiler birds (Shome et al., 2006b), necrotic

JMBFS / Das et al. 2012 : 2 (1) 162-174

164

enteritis in broiler chicken (Das et al., 2008a), acute diarrhea in pigs and piglets (Das et al.,

2009a), haemorrhagic enteritis in elephant and pygmy hog (Das et al., 2008b; Shome et al.,

2010), chronic enteritis in goats (Das et al., 2009b,c), diarrhea in cattle (Das et al., 2012) have

been reported from India. But none of the Clostridial cases from India have been reported

from fish and fish based products.

In literature review, it has been noticed that the research on the involvement of C.

perfringens in freshwater fish is limited (Wen and McClane, 2004; Cai et al., 2008). Equally

limited is the research output on the prevalence of Clostridial infection in commercial fish

based product particularly fish pickle. In light of the limited research available in this field, it

becomes necessary to determine the types of C. perfringens prevalent in fish and fish based

products sourced from Kerala and Tamil Nadu states of India.

Therefore, the present study is designed to identify the prevalence of C. perfringens

types in freshwater fishes and fish pickle packed for human consumption.

MATERIAL AND METHODS

Sample collection

A total of 234 fresh water fish consisting live captured fish (167) consisting common

carps (55), silver carps (56) and cat fishes (56) and stored fish (67) collected from various

dams, freshwater lakes, ponds, retail shops and markets and fish pickles (67) obtained from

randomly selected retail shops and supermarkets of Tamil Nadu and Kerala were acquired for

the study. All the freshwater fish were collected in a regular consumer packages and

immediately transported to the laboratory in an ice chest. Intestinal content samples (234)

from the freshwater fish were then removed scientifically and processed for the

microbiological analysis. Fish pickle were collected in sterilized sample collecting vials and

transported to the laboratory for microbiological analysis.

Isolation and biochemical identification of C. perfringens

Samples collected from the intestinal contents from freshwater fish and fish pickle

samples were inoculated aseptically in Robertson’s cooked meat (RCM) medium (HiMedia,

Mumbai) with neutral oil overlay and incubated at 37ºC for 48 hr. The inoculums from the

JMBFS / Das et al. 2012 : 2 (1) 162-174

165

RCM media was seeded onto 10% sheep blood agar and incubated anaerobically for 24 hr at

37ºC. Bacterial colonies were purified individually based on the size, shape, color, hemolysis

pattern and were subjected to Gram’s and malachite spore staining followed by array of

biochemical tests such as gelatinase, fermentation of glucose, lactose and skim milk and were

identified as per Holt et al. (1994).

Evaluation of phenotypic virulence factors

All the isolates of C. perfringens were evaluated for the deoxyribonuclease (DNase)

enzyme, lecithinase enzyme and hemolysin production as per the protocol described in

Bergey’s manual of determinative bacteriology (Holt et al., 1994) with little modification.

DNase enzyme

The extracellular DNase enzyme was detected on DNase test agar with toludine blue

medium (HiMedia, Mumbai) under aseptic condition. The cultures were spot inoculated into

the media plates and incubated anaerobically at 37ºC for 24 hr. The clear white zones

surrounding the bacterial colonies on a blue colored DNase test medium after 24 hr of

incubation was considered as DNase test positive.

Lecithinase enzyme

The lecithinase activity of C. perfringens was carried out on egg yolk agar base

medium (HiMedia, Mumbai) supplemented with 10 % egg yolk emulsion (HiMedia, Mumbai)

under aseptic condition. The cultures were spot inoculated into the media plates. The

yellowish opalescence (Nagler’s reaction) around the bacterial colonies against the colorless

background of egg yolk agar after 24 hr of incubation at 37ºC in anaerobic condition was

considered as lecithinase test positive.

Hemolysin production

The In vitro hemolysin production or hemolytic activity of C. perfringens isolates

were carried out on solid media plates containing nutrient agar (HiMedia, Mumbai) enriched

with 10 % defibrinated blood of sheep. The cultures were streaked into the media plates.

JMBFS / Das et al. 2012 : 2 (1) 162-174

166

Depending on zone of clearance surrounding the bacterial colonies after 24 hr of incubation at

37ºC in anaerobic condition, result was recorded as either or or both the hemolysis.

Genotyping by polymerase chain reaction

The isolates of C. perfringens were tested to detect the alpha toxin gene (cpa), beta

toxin gene (cpb), epsilon toxin gene (etx), iota toxin gene (iA), enterotoxin gene (cpe) and

beta2 toxin gene (cpb2) by PCR assay (Das et al., 2008a).

Preparation of template DNA

A single colony of C. perfringens was pilled from blood agar plate taken in 100µl of

Milli-Q water, gently vortexed and boiled at 100ºC for 10 min in water bath. The cell debris

was removed by centrifugation at 10,000 rpm for 5 min at 4°C and the top clear supernatant

was used as source of template DNA.

PCR assay

The PCR amplification was carried out in iCycler (BioRad, USA) in 25l reaction

volume 12.5l of 2× PCR master mix [4 mM MgCl2; 0.4 mM of each dNTPs (dATP, dCTP,

dGTP, dTTP); 0.5 units / l of Taq DNA polymerase; 150 mM Tris-HCl PCR buffer (pH

8.5)]; desired M of each forward and reverse primers and 2.5l of template DNA. The

details of primer pairs for cpa, cpb, etx, iA, cpe and cpb2 genes were commercially

synthesized (Eurofins, Bangalore) and were shown in Table 1.

JMBFS / Das et al. 2012 : 2 (1) 162-174

167

Table 1 Oligonucleotides and PCR conditions for detection of cpa, cpb, etx, iA, cpe and cpb2

genes of Clostridium perfringens

Toxin

genes

Oligonucleotide sequences

5/→3/

Concentration of

Oligonucleotides

(M) each

Amplicon

Size (bp)

Reference

cpa

For-gctaatgttactgccgttga

Rev-cctctgatacatcgtgtaag

0.50 324

Shome et al.,

(2006a)

cpb For-gcgaatatgctgaatcatcta

Rev-gcaggaacattagtatatcttc

0.50 180

etx For-gcggtgatatccatctattc

Rev-ccacttacttgtcctactaac

0.50 655

iA For-actactctcagacaagacag

Rev-ctttccttctattactatacg

0.34 446

cpe For-ggagatggttggatattagg

Rev-ggaccagcagttgtagata

0.36 233

cpb2 For-agattttaaatatgatcctaacc

Rev-caatacccttcaccaaatactc

0.36 567

After initial denaturation at 94ºC for 4 min, the amplification cycle had denaturation,

annealing and extension at 94ºC, 55ºC and 72ºC for 1 min each respectively. Final extension

was done at 72ºC for 10 min. C. perfringens (MTCC 3296) and Escherichia coli (MTCC

723) were used as positive and negative controls respectively.

Agarose Gel electrophoresis

The PCR amplicons (5µl) were separated by electrophoresis in 1.5% Agarose

(Promega, USA) gel with TAE 1X (Tris-Acetate-EDTA; pH 8.0) running buffer at 60 V. The

gel was stained with 0. 4 µg/ml ethidium bromide (Himedia; Mumbai), visualized and

photographed in gel documentation system (Universal Hood, BIORAD, Italy).

JMBFS / Das et al. 2012 : 2 (1) 162-174

168

RESULTS AND DISCUSSION

Isolation and identification of C. perfringens

Anaerobic culture process yielded pure colonies of alpha and beta haemolysis on

sheep blood agar. The bacterial colonies isolated from the intestinal contents of fresh water

fish were observed to be centrally dense and round colonies with rhizoid periphery. However,

colonies that originated from pickle were having irregular border with glandular texture.

Gram staining showed gram-positive rods, while malachite green staining showed subterminal

oval endospores. The isolates produced stormy fermentation, acidity, reduction and

coagulation in litmus milk, liquefied gelatin and fermented glucose and lactose (Das et al.,

2008a). Upon the detailed microbiological investigation, a total of 82 (27.24%) isolates of

Clostridium perfringens were tentatively identified (Table 2). Out of 234 intestinal contents

collected from the same number of fresh water fish, C. perfringens were isolated from 43

(18.36%) samples, was almost similar to the previous report from China (Cai et al., 2008),

but lesser than the report from America (Wen and Mcclane, 2004). The rate of isolation of C.

perfringens from 67 samples of fish pickles was estimated to be 39 (58.20%), was higher in

number than the fresh water fish (Table 2).

JMBFS / Das et al. 2012 : 2 (1) 162-174

169

Table 2 Sample details and distribution of virulence genes among the isolates of Clostridium

perfringens from Tamil Nadu and Kerala

Sample details / No of samples

tested

No (%) of C.

perfringens isolated

from tested samples

No (%) of C.

perfringens positive for

toxin genes

cpa cpb2

1. Intestinal contents of Fresh

water fish (n=234)

43 (18.36) 43 (100) 29 (67.44)

a. Fresh water live fish

(n=167)

16 (9.58) 16 (100) 8 (50)

Common carps / 55 3 (5.45) 3 0

Silver carps / 56 5 (8.92) 5 1 (20)

Cat fishes / 56 8 (14.28) 8 7 (87.50)

b. Stored fish (n=67) 27 (40.29) 27 (100) 21 (77.77)

2. Fish pickles / 67 39 (58.20) 39 (100) 36 (92.30)

Total (301) 82 (27.24) 82 (100) 65 (79.26)

Legend: n = total no of samples tested

Evaluation of phenotypic virulence factor

The DNase, lecithinase and hemolysis tests are of low cost and could be useful to

evaluate quantitatively the phenotypes of C. perfringens and for the selection of virulent

strains from the samples collected from the fish and fish based products. All the spot

inoculated isolates of C. perfringens were found to produce DNase and lecithinase activity

(Holt et al., 1994; Das et al., 2009a). It has also been observed that the zone of clearance on

DNase test agar and zone of yellowish opalescence on egg yolk agar emulsion increased

almost double of its size after 48 hr of anaerobic incubation at 37°C (Figure 1). C. perfringens

(MTCC 3296) and C. chauvoei used as positive and negative control respectively.

JMBFS / Das et al. 2012 : 2 (1) 162-174

170

On 10 % sheep blood agar, majority of C. perfringens type A isolates produced double

zone of hemolysis (Figure 1), was reported earlier by several researchers (Holt et al., 1994;

Shome et al., 2006a). All the colonies were found to be surrounded by a smaller zone of beta

hemolysis (complete lysis of blood), which were surrounded by a larger zone of alpha

hemolysis (partial lysis of blood). It has also been observed that the hemolysis production

become sharpened and prominent after 3-4 days of incubation at 4-8ºC.

Figure 1 Extra-cellular Dnase, Lecithinase and Hemolysin production test of C. perfringens

type A isolates A: C. perfringens type A isolates positive for Extra-cellular Dnase

B: C. perfringens type A isolates positive for Lecithinase

C: C. perfringens type A isolates positive for Hemolysin

P: C. perfringens (MTCC 3296) as Positive control

N: C. chauvoei as Negative control

Genotyping by PCR

In this study, PCR was employed for genotyping the 82 C. perfringens isolates

obtained from fresh water fishes and fish pickles. In PCR, out of six virulence genes screened,

alpha toxin gene (cpa) of 324bp fragment were detected in all the 82 (100%) isolates.

However, additional beta2 toxin genes (cpb2) of 567bp fragment were detected in 65

(79.26%) isolates originated from fish pickles (36), stored fishes (21), cat fishes (7) and silver

carps (1) (Table 2; Figure 2). None of the C. perfringens isolates were positive for any of the

cpb, etx, iA and cpe toxin genes. PCR assay thus revealed that all the field isolates originated

from freshwater fish and fish pickles were C. perfringens type A.

JMBFS / Das et al. 2012 : 2 (1) 162-174

171

Figure 2 Detection of alpha (cpa) and beta2 toxin (cpb2) gene of C. perfringens by PCR

A: C. perfringens type A isolates positive for 324bp fragment of cpa gene

B: C. perfringens type A isolates positive for 567bp fragment of cpb2 gene

P: C. perfringens (MTCC 3296) as Positive control

N: E. coli (MTCC 723) as Negative control

Lanes 1-9: Representatives of C. perfringens type A field isolates

M: 100bp DNA ladder plus (MBI Fermentas, USA)

In a similar study, Wen and Mcclane, (2004) reported that out of 34 isolates of C.

perfringens from fish samples 31 were positive for type A (only alpha toxin producer) and 3

isolates were type A with additional enterotoxin gene, but none of the type A isolates were

positive for beta 2 toxin gene. However, in another study conducted by Cai et al. (2008), out

of 75 isolates of C. perfringens from fresh water fish, 13 (17.3%) isolates were found positive

for toxin type A, 58 (77.3%) isolates for type C and 4 (5.3%) isolates were type B, and 47

(62.7%) isolates with additional beta 2 toxin gene (seven from type A, two from type B, and

38 from type C). But none of the isolates were found positive for enterotoxin gene. In

congruence to present result, the cpe- negative C. perfringens type A from the clinical cases

JMBFS / Das et al. 2012 : 2 (1) 162-174

172

of gastroenteritis was reported previously (Das et al., 2009b; Das et al., 2012). In this study,

it has also been noted that alpha toxin is the most prevalent toxin among the C. perfringens

type A isolates (Songer and Meer, 1996), irrespective of sources of isolation and

geographical location. From the results and discussions, it could be suggested that the PCR

assay is a suitable diagnostics for the detection of toxin genes and genotyping of C.

perfringens from the fish and fish products. Both alpha toxin and beta 2 toxins of C.

perfringens type A are lethal and thus the detection of these two toxins in fresh water fish and

fish pickles imply high risk of occurrences of clostridial diseases in human.

CONCLUSION

The Robertson’s cooked meat media broth and the nutrient agar media enriched with

10% sheep blood are suitable for the isolation of C. perfringens. The evaluation of

biochemical tests and phenotypic virulence factors are useful for tentative identification of C.

perfringens. The genotyping by PCR suggested that unlike alpha toxin, the beta2 toxin is also

linked to C. perfringens type A in freshwater fish and fish pickles. Since, C. perfringens is

spore forming and both alpha and beta2 toxins can cause diseases in animal and human by

entering via food chain, therefore, consumption of raw fish pickles and uncooked fishes for

human might cause serious clostridial infection.

Acknowledgments: Authors would like to convey thanks Bannariamman Educational Trust,

Principal, Head of the Department, Biotechnology, Bannari Amman Institute of Technology (BIT),

Sathyamangalam for the kind support and Mrs. Reshmi Deb Choudhury Das and Mrs. Sudhashree

Parvati of Department of English, BIT for editing the manuscript.

REFERENCES

CAI, Y. – GAO, J. – WANG, X. – CHAI, T. – ZHANG, X. – DUAN, H. – JIANG, S. –

ZUCKER, B. A. – SCHLENKER, G. 2008. Clostridium perfringens toxin types from

freshwater fishes in one water reservoir of Shandong Province of China, determined by PCR.

In Deutsche Tierarztliche Wochenschrift, vol. 115, 2008, p. 292-294.

DAHDF. 2009. Annual Report: India, 2008-2009. Department of Animal Husbandry Dairying

and Fisheries, Ministry of Agriculture, Government of India. 2009,

JMBFS / Das et al. 2012 : 2 (1) 162-174

173

http://dahd.nic.in/annualreport2008-09/Book%201-100.pdf (Last downloaded on 23rd June

2012).

DAS, A. – MAZUMDER, Y. – DUTTA, B. K. – KUMAR, A. – SELVI, S. 2009a. Diagnosis

of acute diarrhea in pigs and piglets in Meghalaya, India. In Malaysian Journal of

Microbiology. vol. 5, 2009a, p. 38-44.

DAS, A. – MAZUMDER, Y. – DUTTA, B. K. – SHARMA, G. D. – SHOME, B. R. –

BUJARBARUAH, K. M. 2008a. Clostridium perfringens type A from broiler chicken with

necrotic enteritis. In International Journal of Poultry Science, vol. 7, 2008a, p. 601-609

DAS, A. – MAZUMDER, Y. – DUTTA, B. K. – SHOME, B. R. – BUJARBARUAH, K. M.

– KUMAR, R. 2012.Molecular typing of Clostridium perfringens isolated from diarrhoeic

cattle. In Journal of Animal Science Advances, vol. 2, 2012, p. 226-229.

DAS, A. – MAZUMDER, Y. – DUTTA, B. K. – SHOME, B. R. – BUJARBARUAH, K. M.

– SHARMA, G. D. 2008b. Clostridium perfringens type A beta2 toxin in elephant (Elephas

maximus indicus) and pygmy hog (Sus salvanius) with haemorrhagic enteritis in Assam,

India. In African Journal of Microbiology Research, vol. 2, 2008b, p. 196-201.

DAS, A. – MAZUMDER, Y. – RAJESH LAL, L. P. M. 2009b. Diagnosis of chronic enteritis

in goats by polymerase chain reaction. In International Journal of Biotechnology and

Biochemistry, vol. 5, 2009b, p. 27-34.

DAS, A. – RAJESH LAL, L. P. M. SENTHAMARAI, M. – SENTHILRAJA, K. 2009c.

Polymerase chain reaction in diagnosis of chronic enteritis in goats. In Research Journal of

Biotechnology, vol. 4, 2009c, p. 25-29.

FAOUN. 2011. Fishery and Aquaculture Country Profiles: India. Food and Agriculture

Organization of the United Nations. 2011, http://www.fao.org/fishery/countrysector/FI-

CP_IN/3/en (Last downloaded on 23rd June 2012)

HOLT, J. G. – KRIGE, N. R. – SNEETH, P. H. A. – STALEY, J. T. – WILLIAMS, S. T.

1994. Bergey’s Manual of Determinative Bacteriology. Ninth edit. Williams and Wilkins

Baltimore, USA, 1994, 749-755 p. ISBN 0683006037 9780683006032.

SHOME, B. R. – SHOME, R. – BUJARBARUAH, K. M. – DAS, A. – RAHMAN, H. –

SHARMA, G. D. - DUTTA, B. K. 2010. Investigation of haemorrhagic enteritis in pygmy

hogs (Sus salvanius) from India. In Scientific and Technical Review - International Office of

Epizootics, vol. 29, 2010, p. 687-693.

SHOME, B. R. – SONGER, J. G. – SHAKUNTALA, I. - SHOME, R. – KUMAR, A. –

CHAKRABORTY, S. – MAZUMDER, Y. – RAHMAN, M. M. – DAS, A. – MURUGKAR,

H. V. – RAHMAN, H. – BUJARBARUAH, K. M. 2006a. Atypical blackleg caused by

JMBFS / Das et al. 2012 : 2 (1) 162-174

174

Clostridium perfringens Type A in cattle in Manipur, India. In Indian Journal of Animal

Science, vol. 76, 2006, p. 353-357.

SHOME, R. – SHOME, B. R. – DAS, A. – KUMAR, A. – RAHMAN, H. –

BUJARBARUAH, K. M. 2006b. Characterization of Clostridium perfringens associated with

gangrenous dermatitis in broiler birds and confirmation by PCR. In Indian Journal of Poultry

Science, vol. 41, 2006b, p. 218-220.

SMEDLEY III, J. G. – FISHER, D. J. – SAYEED, S. – CHAKRABARTI, G. – MCCLANE,

B. A. 2004. The enteric toxins of Clostridium perfringens. In Reviews of Physiology,

Biochemistry and Pharmacology, vol. 152, 2004, p. 183-204.

SONGER, J. G. – MEER, R. R. 1996. Genotyping of Clostridium perfringens by polymerase

chain reaction is a useful adjunct to diagnosis of clostridial enteric disease in animals. In

Anaerobe, vol. 2, 1996, p. 197-203.

SONGER, J. G. 1996. Clostridial enteric diseases of domestic animals. In Clinical

Microbiology Review, vol. 9, 1996, p. 216-234.

STING, R. 2009. Detection of beta2 and major toxin genes by PCR in Clostridium

perfringens field isolates of domestic animals suffering from enteritis or enterotoxaemia. In

Berliner und Munchener tierarztliche Wochenschrift, vol. 122, 2009, p. 341-347.

WEN, Q. – MCCLANE, B. A. 2004. Detection of enterotoxigenic Clostridium perfringens

Type A isolates in American retail foods. In Applied and Environmental Microbiology, vol.

70, 2004, p. 2685–2691.

YAMAGISHI, T. – SUGITANI, K. – TANISHIMA, K. – NAKAMURA, S. 1997.

Polymerase chain reaction test for differentiation of five toxin types of Clostridium

perfringens. In Microbiology and Immunology, vol. 41, 1997, p. 295-299.